Transcript Contrast Induced Acute Kidney Injury

Contrast Induced Acute
Kidney Injury
Cardiology Catheterization
Conference
4/14/2011
Peter Danyi MD
The scope

Pubmed search
• Contrast induced nephropathy-4671
• 718 reviews
• First report 1955, Alwall et al. Acta Med
Scand
Definition




An increase in serum creatinine generally occuring within
24hrs, peaking up to 5 days after, and returning to baseline
within 3 weeks.
Definition varies, increase in Cr of >25-50% from baseline
and/or >0.5-1.0 mg/dL after 48-72 hrs.
European Society of Urogenital Radiology 25% or 0.5 mg/dl
Second International Consensus Conference of the Acute
Dialysis Quality Initiative – defines renal failure in terms of
GFR and urine output.
Epidemiology
CIN Consensus Working panel report 2006



First reports over 50 years ago
Common and potenitally serious complication following
administration of constrast media (CM) in patients with risk
for acute renal injury
Incidence reports varies widely depending on patient
population and baseline risk factors
Definition and frequency of CIN
The frequency of contrast-induced nephropathy depends on the definition used. N 1,826 patients
undergoing percutaneous coronary intervention. (Adapted from Am J Med3 and J Invasive
Cardiol.4)
Definition in number of trials
Numbers of trials in the literature using the following definitions of contrast-induced
nephropathy: 0.5 mg/dL (44.2 mol/L) increase, 1.0 mg/dL (88.4 mol/L) increase, 25% increase,
Incidence


Large scale studies in general hospital patients provide best
estimate of healthcare impact
CIN has decreased over the past decade from ~15% to
~7% recently
• Nash et al. 4622 patients, 7.2% CIN

Thought to be due to
• Greater awareness of the problem
• Better risk prevention measures
• Improved iodinated contrast media

Total number of cases did not decline due to increasing
number of procedures requiring contrast
Impact




Radiographic contrast media is the third most common
cause of hospital acquired renal failure (after decreased
renal perfusion and nephrotoxic medications), accounting
for 11% of cases
In-hospital mortality rate of CIN as high as 14%1
In patients with multiple risk factors, incidence of CIN can
rise to 50% or great er2
Differential diagnosis: cholesterol embolism-can be difficult
to distinguish, but typically occurs over 3-8 weeks, in
elderly men, may have cutaneous signs
1 Nash et al. Am J Kidney Dis 2002
2 McCullough PA et al. Am J Cardiol 2006
Mortality risk





Levy et al., 1996, retrospective study of >16,000 patients
with 2% CIN incidence, but 34% vs. 7% risk of death for
CIN
After adjustment the odds of dying was 5.5 for patient with
CIN
Rihal et al., 2002, retrospective study of 7,586 patients
noted a CIN incidence of 3.3% and a 22% vs. 1.4% inhospital mortality for patients with CIN
McCullough et al., 1997 reported a mortality rate of 7.1%
for CIN, 35.7% if patient required HD, vs. 1.1% for no CIN.
2 year mortality was 81.2% for HD patients
Significant association was found between >25% Cr
increase and one year mortality (Gruberg et al, 2000)
Creatinine rise and1-year mortality
Depiction of 1-year mortality rates according to postprocedure increase in serum creatinine J
Am Coll Cardiol
Clinical course and outcome

CIN is associated with late cardiovascular events
• Lindsay et al. 2003: 5,967 patients, post PCI registry showed
AMI 24% vs. 11.6%; TVR at 1 year 28.8% vs. 20.3%
• Lindsay et al. 2004: CIN is associated with CKMB rise, CIN and
CKMB rise together has 26% 1 year risk for death or MI

Higher incidence of in-hospital events with CIN
•
•
•
•

CABG
Bleeding requiring transfusion
Vascular complications
Longer hospital stay
Gruberg et al. did not find increased MI or revascularization
rates
Economic impact




Subramanian et al. in 2007 published an economic analysis
in J. Med. Econ. of CIN
Average additional cost was $10,345 for the hospital stay
and $11,812 to 1 year
Incidence and outcome data was determined through
systematic literature search
The major driver for the cost was increased hospital stay
Pathophysiology




Poorly understood, complex syndrome, little known about
underlying cellular mechanism
Available contrast media constitute diverse family (ionic,
nonionic, high/low/iso-osmolar)
Biphasic response or immediate vasoconstriction (dose
dependent) and reduction of renal blood flow (up to 50%,
lasting for hours)
Subsequent stasis of contrast causes
• direct tubular injury and death (osmotic nephrosis-intense
focal or diffuse vacuolization of the proximal tubules- or overt
tubular necrosis))
• vasoconstriction through Ca influx causing regional corticomedullary and outer medullary blood flow reduction
Pathophysiology



Endothelins (if contrast >150 ml), adenosine (prolonged
vasoconstriction, A1 receptors) and angiotensin II released
causing vasoconstriction
Nitric oxid production reduced (in proportion to osmolality
of agent), reactive oxygen species production (due to
hypoperfusion) increased
Iron plays a role in oxidative injury in CIN - ? role of iron
chelation therapy
Pathophysiology


Chronic Kidney Disease (CKD) predisposes to the
development of contrast-induced acute kidney injury (AKI).
Contrast-induced AKI assumes baseline reduced nephron
number with superimposed acute vasoconstriction
Any superimposed insult such as sustained hypotension,
microshowers of atheroembolic material from catheter
exchanges, use of IABP, or a bleeding complication can
amplify the injury processes occurring in the kidney
Factors involved in CIN
pathogenesis
Postulated Pathophysiology of CIN
In the presence of a reduced nephron mass, the remaining nephrons are vulnerable to injury. Iodinated contrast, after causing a brief
(minutes) period of vasodilation, causes sustained (hours to days) intrarenal vasoconstriction and ischemic injury. The ischemic injury sets
off a cascade of events largely driven by oxidative injury causing death of renal tubular cells. If a sufficient mass of nephron units are
affected, then a recognizable rise in serum creatinine will occur
Risk factors
Other risk factors

Increasing age

Gender data inconsistent



Earlier studies reported higher incidence in nonwhites
(Lindsay, 2003) and African Americans (Dangas, 2005) but
these findings have not been replicated
Intraarterial administration and osmolality and volume of
contrast
The effect of risk factors is additive. If a patient has 4 or 5
risk factors, risk of CIN is 50%
Risk models
McCullough et al. Am J Cardiol 2006
CIN Mehran Risk Score
.
Mehran et al. JACC 2004
Risk reduction strategies













Contrast
IVF
NaHCO3
N-Acetylcysteine
Ascorbic acid
Statin
ACEi
Theophyllin/aminophylline
Diuresis
Dopamine/Fenoldopam
Prostaglandin/prostacyclin
ANP
Mechanical (HD, hemofiltration, RenalGuard)
Contrast


First contrast agent described in CIN was iodopyracet, a diiodinated pyridine derivative. That class was replaced by
tri-iodinated benzoic acid derivatives and later high osmolar
contrast media (HOCM), 2000 mOsm/kg
HOCM is upto 8 times the plasma osmolality, includes:
• Diatrozate, Metrizoate, Ioxithalamate, Iothalamate


Osmolality can be reduced by less iodine in the molecule or
using nonionic agents (LOCM), 6-800 mOsm/kg. Nonionic
monomers: iohexol (Omnipaque), iopamidol, iopentol, iomeprol,
iobitridol, ioversol. Dimer: ioxaglate
Iso-osmolar nonionic dimer, 290 mOsm/kg: iodixanol
(visipaque)
Contrast



31 studies evaluate HOCM vs LOCM, pooled data shows
LOCM associated with significantly less CIN, pooled OR 0.61
(0.48-0.77)
Comparison of LOCM-s: small studies, no significant
difference overall. 64% of patients had some increase in Cr
IOCM (iodixanol) and LOCM have been compared, all in
2009:
• Wessely et al. (N=324) did not found a difference (iomeprol)
• Chuang et al (N=50) did not found a difference (iohexol)
• Mehran et al. ICON study (N=146) nonsignificant difference
(ioxaglate)
• Laskey et al. (418 DM pts) no difference (iopamidol-Isovue)
Contrast




LOCM vs. IOCM- meta-analysis of 16 studies with 2763
patients
No difference when IOCM compared to LOCM collectively
Iodixanol (IOCM) had lower incidence of CIN when
compared to ioxaglate and iohexol (Omnipaque) separately
No difference when compared to other agents
Contrast



Contrast volume is independent predictor for CIN, although
not all studies have shown that. The worse the CKD is, the
lower amount of contrast can cause CIN.
CIN consensus panel: >100 ml of contrast is associated
with higher incidence of CIN, but there is no threshold
amount
Intra-arterial administration is associated with higher rate
of CIN and LOCM is more beneficial over HOCM than with
intravenous use
Statins


Pleiotropic effect of antioxidative and anti-inflammatory
properties
Systematic review (Xhang, Am J Nephrol, 2011) found 6
cohort studies and 6 RCTs. Heterogeneity found among
studies.

4/6 cohort studies found chronic statin therapy beneficial

Most RCTs failed to show benefit.


Chronic statin therapy may be more beneficial than only
around fhe time of CM administration
Dose of beneficial statin uncertain
Statins
Zhang et al. Am J Nephrol. 2011
Statins
Zhang et al. Am J Nephrol. 2011
Statins-RCTs
N-Acetylcysteine




Scavanges ROS, reduces the depletion of glutathione,
stimulate vasodilatory mediator release (incl. nitric oxid)
Tepel et al. (NEJM 2000) first described its efficacy in
preventing CIN
Since then a multitude of trials published with highly
conflicting results (largest trial 487 patients)
A number of meta-analysis published
• Kelly et al. analyzed 26 trials, 3393 pts, RR=0.62 (0.44-0.88)
• Gonzales et al. 22 trials, 2746 pts, found two clusters- one
NAC was protective, Cr decreased-NAC not actually effective,
but rather an artifactual Cr decrease not due to NAC
• Trivedi et al. analyzed high dose NAC only, OR=0.46 (0.330.63)
N-Acetylcysteine
N-Acetylcysteine
Kelly et al. Ann Int Med 2008;148:284-94
IV Fluids

IVF protects against CIN in two ways
• Intravascular space expansion blunts the vasoconstrictive
effect of contrast in the medulla by suppressing vasopressin,
RAAS, increasing prostaglandins
• Attenuates direct toxic effect of contrast on tubules by
decreasing concentration and viscosity



Solomon et al (1994 NEJM) published that 0.45% NS was
more effective in preventing CIN than IV mannitol or
furosemide (11% vs 28% vs 40%)
Trivedi et al (Nephron 2003) confirmed that IVF itself was
beneficial (CIN 3.7% for 0.9% NS vs. 34.6% unrestricted
oral fluids)
The effect of tonicity of IVF was evaluated by Mueller et al.
(Arch Int Med 2002) low risk pts randomized to 0.45 vs 0.9
NS. (CIN was 2% vs. 0.9%)
IV fluids

More recently isotonic normal saline vs sodium bicarbonate
has been the focus of investigation

The generation of ROS thought to be mitigated by NaHCO3

10 clinical trials in the past 5 years (largest trial 502
patients)


6 had lower incidence of CIN in the NaHCO3 group, 4
showed no significant benefits
11 meta-analysis published
• Zoungas et al. (Ann Int Med 2009) 23 pooled studies RR 0.62
(0.45-0.86), but found significant publication bias, unpublished
trials did not show benefit
• Kunadian et al. (Eur J Radiol 2010) found the same
NaHCO3 vs. NaCl
Weisbord, Pavelsky. Curr Opin Nephrol Hypertens. 2010
Atrial natriuretic peptid

Morikawa et al (JACC 2009) single center RCT ANP+IVF or
IVF alone cath+/-PCI

ANP at 0.042 mcg/kg/min 4-6 hrs prior and 48 hrs past

eGFR at 24, 48 hrs, 1 week and 1 month

Slight benefit of ANP in Cr, no benefit in HD/hosp. rate


Prior studies were negative, but this on had lower ANP dose
for longer time
No conclusive evidence yet, would need larger trials
Prostoglandin/prostacyclin analogs






PGI2, PGE2 have renal vasodilatory effects (this forms the
basis of NSAIDs discontinuation)
Animal models suggest a potential benefit
Spargias et al. (Circ. 2009) RCT 208 patients –iloprost
(PGI2 analog) found CIN 8% vs. 22%, p=0.005
No longer term effect investigated
Study powered for 70% reduction in CIN-clinically
implausible
Large clinical trials needed before able to recommend
Renal perfusion and diuresis
Ascorbic acid, an antioxidant was found to be associated with
less CIN in 2-3 gm/day doses before and after the procedure in
one study (Spargias et al. Circ. 2004)
Calcium channel blockers were found to have no effect on the
incidence of CIN
Theophylline
Stacul et al. (Am J Cardiol, 2006) found significant benefit of
adenosine antagonist administration (theophylline and
aminophylline)
Kelly et al. Ann Intern Med. 2008
ACE inhibitors





ACEi/ARB play an integral role in cardiac patient care, DM
nephropathy prevention, AMI, HTN
Angiotensin II is a potent vasoconstrictor, stimulator of
vasopressin and aldosterone
Xiong et al. (Pharmacol Res 2006) demonstrated that
irbesartan attenuates CIN renal tubular cell apoptosis by
reducing oxidative stress
Gupta et al. (Indian Heart J 1999) showed that captorpil
reduced CIN in DM patients by 79%
Dangas et al. (AM J Cardiol 2005) ACEi reduces risk of CIN
in CKD patients
ACE inhibitors



Holscher et al. in a 412 pt study found ACEi associated with
CIN (Can J Cardiol 2008), Toprak et al. found similar results
in 2003
Cirit et al. evaluated chronic ACEi users and found that ACEi
increases the likelihood of CIN, likely due to decrease in
glomerular hydrostatic pressure
Rosenstock et al. published the largest RCT to date on ACEi
and CIN. 283 patients, three groups (ACEi cont, ACEi
discont, ACEi naive). IVF and NAC were used. Found no
significant difference.
Available studies for ACEi and CIN
Patel K, King C, Jovin IS Cardiovasc Revasc Med 2011, 90-93
McGuire VAMC study






2201 mostly (98%) male patients, retrospectively
analyzed from Janury 2003 until September
2005.
Average follow up 65 months.
Risk factors, interventions collected
We were interested if IVF administration, ACEi,
NAC or statin therapy has any benefit on CIN and
beyond the traditional definition (at 3 months and
for risk of HD)
We included diagnostic cardiac catheterization
and PCI/PEI
Included patients with all creatinine levels
McGuire VAMC study
McGuire VAMC study-IVF
McGuire VAMC study -ACEi
McGuire VAMC study - statin



CIN occurred in 3% patients in the statin group and 3.6%
in the no statin group at 72 hours after the procedure
OR 0.82 (0.43 – 1.55) P=0.54
At 3 months, CIN was seen in 7.8% patients of the statin
group versus 7.9% of the no statin group OR 0.99 (0.64 –
1.51) P=0.95
The Cox proportional hazard ratio for starting HD in people
on statin at the time of catheteriation was 0.88 (0.44-1.77)
p=0.72
McGuire VAMC study-Mucomyst




The NAC treated group was significantly older, had
significantly more CAD, CHF, DM, and higher baseline
creatinine (mean 1.75 vs. 1.20)
CIN occurred in 5.6% patients in the NAC group and in
2.7% in the no NAC group at 72 hours after the procedure
OR 2.14 (1.08 – 4.23)
At 3 months, CIN was seen in 13.5% patients of the NAC
group versus 6.8% of the no NAC group OR 2.14 (1.36-3.3)
p=0.0008
NAC was administered to a significantly sicker population
with a much higher risk of CIN development, but NAC
preventitive therapy did not seem to have a protective
effect
General limitations in the literature





Most studies without placebo/control group
Different baseline characteristics among studies-different
risk for CIN/CIAKI
Different interventions for prevention (different IVFs with or
without NAC or other interventions) not consistent among
studies
Different CM administration (cardiac cath, +/- PCI, IV
contrast)
Studies have limited number of patients
Stacul et al. Am J Cardiol 2006
Mechanical interventions





Rapid removal of contrast media from circulation would
limit glomerular damage, decrease risk.
Conventional HD showed no benefit in 5 studies and there
was a trend toward harm
Marenzi et al. compared compared continuous
hemofiltration initiated 4-8 hrs before coronary angiography
and continuing 18-24 hrs after. CIN was 5% versus 50%.
However, since CVVH directly lowers Cr, it is felt that most
of the benefit was due to that.
Cruz et al. (2006) conducted a meta-analysis of blood
purification therapies (6 HD, 2 CVVH) no benefit.
Holscher et al (2008), RCT (N=412) found potential harm
when compared to NAC (16.4% vs. 4.3%) and IVF (16.4%
vs. 7.2%)
Mechanical interventions

RenalGuard
• PRINCE and REMEDIAL II trial
• Creates and maintains high urine output with loop diuretic and
replaces it real time with equal amount of normal saline
• Matched fluid replacement minimizes risk of over- or
underhydration

Coronary sinus withdrawal of contrast
• Venous contrast capture (VCC) by balloon tipped catheter
• Safely removes ~70% of contrast (Meyer et al. J Interv Cardiol
2006)
• Danenberg et al.: ~44% of contrast- Cardiovasc. Revasc. Med
2008
Novel approaches


Large studies of oral/iv antioxidants (deferiprone)
Intrarenal infusions of renal vasodilators using flow directed
catheters

Systemic Cooling

Novel, less toxic forms of radio-opaque contrast agents
Suggested preventive interventions
Weisbord, Pavelsky. Curr Opin Nephrol Hypertens. 2010
Conclusions



Although there is evidence that CIN is associated with
worse outcomes, there is no evidence of a causal
relationship or that reducing CIN improves overall outcome
There is a large body of literature, but inconsistent inclusion
criteria, definitions and conflicting results
To date there has been no conclusive benefit demonstrated
from interventions except for low osmolar or iso-osmolar
contrast and IV hydration. These showed benefits mainly in
the short term.
Conclusions



In case CIN is causative for worse long term outcome, it is
wise to reduce the number of studies requiring contrast, the
amount of contrast administered
IVF administration for adequate hydration is helpful at least
in the short term, if CHF is not present there is little
downside
NAC data is inconclusive, but it is cheap, easy to administer
and with minimal side effects. It is recommended to use in
high dose.